Forensic Autopsies can Determine Latent Prostate Cancer Prevalence

Abstract

To evaluate the usefulness of forensic autopsies in determining latent prostate cancer (PC) prevalence, we examined latent PC prevalence from autopsies and compared our findings between decedents with and without cancer. Data from forensic autopsies performed in Japan from 2004 to 2014 were obtained. For each prostate, histopathological examinations were performed in both the base and the apex sections. Three hundred and seventeen Japanese decedents were selected for analysis. The mean age of decedents was 56.4 ± 17.8 years (range, 14–94 years). Among this population, 39.4% died suddenly of disease and 60.6% died of external causes. Latent PC was identified in 45 (14.2%) decedents, who ranged from 27 to 93 years old (mean, 71.1 ± 12.9 years). The prevalence of clinically significant PC with a Gleason score of 7 or more was 8.8%, and the rate increased with age. Fifteen males had cancers other than PC. The prevalence of overall latent PC was significantly higher for those with cancer compared with those without (40.0% vs. 12.9%; P = 0.003). In this study, the use of forensic autopsy materials provided the opportunity to obtain a more accurate natural history of PC, as the decedents in this situation would have been more likely to have died suddenly while behaving as normal prior to death, and less likely to have been impacted by long‐term medical interventions.

Low‐risk prostate lesions: An evidence review to inform discussion on losing the “cancer” label

“…All studies were rated at a high risk of bias due to non‐representativeness of the study sample or to minimal sampling of prostate tissue (Supporting Information: 4b ). The overlapping studies by Kido et al 57 and Takaso et al 58 describe forensic autopsies performed at Dokkyo Medical University, Japan from August 2002 to July 2005 ( n = 196), and November 2004 to February 2014 ( n = 317), respectively. In both studies, the prostate gland was cut into 3 vertical slices, and only 2 cross‐sections were examined.…”

“…All studies were rated at a high risk of bias due to non-representativeness of the study sample or to minimal sampling of prostate tissue (Supporting Information: 4b). The overlapping studies by Kido et al 57 and Takaso et al 58 3)…”